The first randomized controlled trial of a GLP-1 drug for alcohol use disorder reports that semaglutide cut heavy drinking days by 13.7 percentage points more than placebo over twenty-six weeks. The trial enrolled 108 adults at a Copenhagen mental-health center. All of them also had obesity. The Lancet published the paper on May 2. [1] The clinical pharmacology side of the GLP-1 class — including the intracellular-activation variant the paper covered yesterday — is moving in parallel.

The mechanics are tighter than the headline suggests. Mette Kruse Klausen of Copenhagen University Hospital was first author; the senior corresponding author was Anders Fink-Jensen. The co-author list includes George F. Koob, director of the U.S. National Institute on Alcohol Abuse and Alcoholism, and Nora Volkow, director of the National Institute on Drug Abuse. [1] Participants were randomized to once-weekly subcutaneous semaglutide titrated up to 2.4 mg, or matching saline placebo. Both arms received up to ten standardized cognitive behavioral therapy sessions for alcohol use disorder. At baseline, participants reported seventeen heavy drinking days in the prior thirty. At twenty-six weeks, the semaglutide arm had reduced that number to roughly five; the placebo arm to nine. Total alcohol intake fell from around 2,200 grams per thirty days to about 650 grams in the semaglutide group, and to about 1,175 grams in placebo. [2]

The treatment difference of 13.7 percentage points (95% CI, -22.0 to -5.4) is the primary endpoint. [1] Semaglutide also outperformed placebo on plasma γ-glutamyl transferase and phosphatidyl ethanol, biomarkers that corroborate the self-reported drinking data. The trial's number needed to treat — a clinical-efficacy metric — was 4.3, compared with seven or higher for the three medications currently approved for alcohol use disorder. [3] Koob, in the NIH research-matters write-up, called a more accessible option for AUD "a gamechanger for closing the treatment gap." [3] About 28 million American adults meet the criteria for alcohol use disorder; only a small minority receive any pharmacotherapy.

The limits are important. The trial was small, single-center, and Danish. It enrolled only people with body mass index of thirty or higher, which means the findings do not necessarily generalize to AUD patients without obesity. There was no follow-up after the twenty-six-week treatment period, so the durability of the reduction is unknown. People who lose weight on GLP-1 drugs frequently regain it after discontinuation; an analogous bounce in alcohol consumption is plausible, given GLP-1 mechanisms work through brain reward circuitry. Matt Field of the University of Sheffield, in a Science Media Centre reaction, framed it as "the strongest evidence yet that these medications may help some people to reduce their alcohol consumption" while noting that the trial does not tell us what happens when patients stop. [4] A companion Lancet commentary by Christian Hendershot and Klara Klein called for larger trials in more diverse populations before this can inform routine clinical practice. [1]

The trial widens the GLP-1 indication map. Semaglutide is approved for diabetes and obesity; the paper's Wednesday piece on intracellular GLP-1 activation covered the next-generation chemistry. Observational studies have suggested benefits in sleep apnea, cardiovascular outcomes, and mental health. This is the first randomized prospective evidence — gold-standard in clinical research — that a GLP-1 drug can change drinking behavior. Off-label prescribing for AUD has already started; the trial gives it a formal evidence base, while leaving the FDA-desk question of whether to pursue a labeled indication open. Wegovy and Ozempic, the two semaglutide formulations on the market, are not approved for alcohol use disorder, and Novo Nordisk has not publicly announced plans to file.

What the trial gives clinicians is a different conversation. A patient with comorbid obesity and AUD who is already considering semaglutide for weight has, as of May 2, a peer-reviewed Lancet paper suggesting the drug may help with both. Twenty-eight million Americans is a denominator that justifies the next, larger trial.

-- NORA WHITFIELD, Chicago